commit | cb74476fe47e16467c09308d8a3a10850829bd9c | [log] [tgz] |
---|---|---|
author | Kunwar Grover <groverkss@gmail.com> | Fri Sep 06 13:49:23 2024 +0100 |
committer | GitHub <noreply@github.com> | Fri Sep 06 13:49:23 2024 +0100 |
tree | 8b72005735951e041e12490d37fb73ef6680cd5d | |
parent | 767e2881bd90aa3f8aa515bae7bf75602fc4f355 [diff] |
[Codegen][GPU] Add gpu configuration heuristics for attention (#18415) MMA heuristics are refactored into utilities, and these utilities are used to generate a schedule for attention. For now, attention still uses the same heuristic as GEMM to calculate required tile sizes. This heuristic will be improved in future (the heuristic for attention is to have as many M tiles as possible). This does not modify any tests, as MMA configuration logic is same. Attention configuration logic is also same, but now has more checks for shared memory and alignment for the QK matmul.
IREE (Intermediate Representation Execution Environment, pronounced as “eerie”) is an MLIR-based end-to-end compiler and runtime that lowers Machine Learning (ML) models to a unified IR that scales up to meet the needs of the datacenter and down to satisfy the constraints and special considerations of mobile and edge deployments.
See our website for project details, user guides, and instructions on building from source.
IREE is still in its early phase. We have settled down on the overarching infrastructure and are actively improving various software components as well as project logistics. It is still quite far from ready for everyday use and is made available without any support at the moment. With that said, we welcome any kind of feedback on any communication channels
Package | Release status |
---|---|
GitHub release (stable) | |
GitHub release (nightly) | |
Python iree-compiler | |
Python iree-runtime |
Host platform | Build status |
---|---|
Linux | |
macOS | |
Windows |
For the full list of workflows see https://iree.dev/developers/general/github-actions/.
See our website for more information.
Community meeting recordings: IREE YouTube channel
IREE is licensed under the terms of the Apache 2.0 License with LLVM Exceptions. See LICENSE for more information.